Are we butterflies or computer games?

Elon Musk says we are almost certain to be computer simulations, but a new paper argues we are statistically more likely to be living in the real world, after all

Alex Lo
Published:  28 Sep, 2020

“Once Zhuang Zhou dreamt he was a butterfly, a butterfly flitting and fluttering around, happy with himself and doing as he pleased. He didn’t know he was Zhuang Zhou. Suddenly he woke up, and there he was, solid and unmistakable Zhuang Zhou. But he didn’t know if he was Zhuang Zhou who had dreamt he was a butterfly or a butterfly dreaming he was Zhuang Zhou.”

– the Zhuangzi

We are all nothing more than a computer game, according to tech billionaire Elon Musk. As he famously put it during a conference in 2016, “There’s a one in billions chance we’re in base reality.”

Personally, I prefer being a butterfly than a computer simulation. But with that kind of probability, can’t we just say with almost certainty that we are nothing more than computer sims? Is that why Musk often says strange things and behaves strangely? If you are convinced your life is just a clever lifelike simulation, would you behave differently from before? Or would that be part of your programmed awareness as well?

But before you go and make any big life changes, you may want to know someone has come up with a credible and very clever calculation that significantly – seriously! – lowers the chances of our being simulations to less than 50 per cent. Well, still quite high, still very disturbing, but a lot better than Musk’s odds!


The paper, “A Bayesian approach to the simulation argument”, is by David Kipping, an astronomer at Columbia University, that was published last month in the MDPI journals.

He wrote: “The probability that we are sims is in fact less than 50 per cent, tending towards that value in the limit of an infinite number of simulations.”

“An infinite number of simulations” refers to Musk’s billions of simulated worlds. Musk’s argument is in turn based on the famous paper, “Are You Living in a Computer Simulation?”, published by Oxford philosopher Nick Bostrom in 2003, which started the current controversy among brainy geeks and fans of the Matrix sci-fi series.

Bostrom presents three possibilities: (1) humans will never achieve a simulated reality; (2) we achieve the technology but don’t use it; (3) we achieve and use it.

We can probably all agree that (2) is highly unlikely. We humans have always used whatever technology invented, however terrible, like the atomic bomb. Musk provides arguments against (1) and for (3).

Against (1), he says: “The strongest argument for us being in a simulation, probably being in a simulation is the following: 40 years ago, we had pong, two rectangles and a dot,” Musk said. “That is what games were. Now 40 years later we have photorealistic 3D simulations with millions of people playing simultaneously and it’s getting better every year. And soon we’ll have virtual reality, augmented reality, if you assume any rate of improvement at all, the games will become indistinguishable from reality.”


Actually, the strongest argument – that’s Musk’s “one in billions” – lies with (3). Once beings in the “real” world or what these guys call “base reality”, have achieved true simulation tech, it will spawn not one but many, which in turn will create more and more sims, possibly at an exponential rate. Hence, Musk’s “one in billions” and Kipping’s “tending towards that value in the limit of an infinite number of simulations”.

In other words, you and I, statistically, are far more likely to inhabit one of those billions of sims than the “one and only” base reality/real world.

“If post-human civilisations eventually have both the capability and desire to generate such Bostrom-like simulations,” Kipping wrote, summarising his opponents’ argument, “then the number of simulated realities would greatly exceed the one base reality, ostensibly indicating a high probability that we do not live in said base reality”.

Bayesian statistics allows for the updating of new information to determine the probability based on the latest input.

Kipping’s Bayesian approach allows him to calculate the probability of us being sims when we have not yet mastered or used the technology – Bostrom’s (1) and (2), which is also the situation we are presumably in now – as well as the probability of us being sims once that technology becomes available and is used.

It’s interesting to note that Kipping makes a clever use of Descartes’ cogito, ergo sum in his calculations. That’s fitting, since Descartes’ “evil genius” thought experiment may be said to be the precursor of the sims argument. Kipping uses “I think, therefore I am” as the single one known datum that must exist in all and every possible simulated world as well as base reality.

His result, based on (1) and (2), as we report at the start, is that our chances of being sims are less than 50 per cent. But once we have achieved (3), that probability is reversed, meaning there is then a more than 50 per cent chance of us being sims.

Kipping’s odds are less disturbing than those of Musk and Bostrom, but hardly reassuring, that is, if you prefer to live in the real world. Perhaps this can be a Stephen King-type horror movie. Imagine you are a reporter interviewing a group of scientists who have just achieved a computer simulation powerful enough to resemble real life.

You really need to pause – because your chances of being a sim yourself in a computer programme has just risen considerably.

From all these readings, and you may add Plato’s famous allegory of the cave in the Republic, the Matrix movies and any number of postmodern theories that deny “reality” as such, you may think the West really suffers from an extreme case of epistemological anxiety or anguish. Consider those images: prisoners chained up inside a dark cave; an evil genius who uses “all his time, powers and energies” to deceive Descartes; humans kept alive in capsules and used as batteries …

Many people, over the ages, have compared the famous butterfly dream of the Zhuangzi to Plato’s cave and Descartes’s demon. Thematically, they are certainly comparable, but their literary tones are completely different. In the iconic classical Chinese planting, Zhuang Zhou is portrayed in a serene repose, dreaming of being a butterfly dreaming of being Zhuang Zhou.

Zhuang Zhou or the butterfly is just having a pleasant dream.